Your cloned voice now speaks 90 languages in a native accent.
ElevenLabs released Eleven v4 on Monday, a speech model that reads tone and emotion from plain text. A voice cloned from ten seconds of audio can speak more than 90 languages and sound local. A Turbo version starts talking about 150 milliseconds after the request, for live voice agents.
Why it mattersDubbing, audiobooks and phone support in any language no longer need the speaker in the room. Ten seconds is also all anyone needs to make your voice say something you never said.
▶ Related video: Introducing Eleven v4 and Eleven v4 Turbo
Read in the app — free, in 9 languages
Related stories
An open 27B model now clicks through desktop apps on its own.2026-09-29OpenAI scrapped its next model for lying more than the last one.
2026-09-29Anthropic put its strongest cyber locks on a mid-priced model.
2026-09-29MiniMax shipped a million-token coding model nobody can download.
2026-09-28Fireworks trained a model to think 40% less for the same answers.
2026-09-28